Title :
Landmine detection in high resolution 3D GPR images
Author :
Ligthart, E.E. ; Yarovoy, A.G. ; Roth, F. ; Ligthart, L.P.
Author_Institution :
Int. Res. Center for Telecommun.-transmission & radar, Delft Univ. of Technol., Netherlands
Abstract :
This work describes a novel landmine detection and classification algorithm for high resolution 3D ground penetrating radar (GPR) images. The algorithm was tested on data measured with a video impulse radar (VIR) system developed by the International Research Centre for Telecommunications-transmission and Radar (IRCTR). The algorithm detected all landmines (including the difficult to detect M14 mines) and classified almost all specific landmines features correctly with a reasonably low false alarm rate. The feature on a small height object demonstrates an effective reduction of false alarms.
Keywords :
feature extraction; ground penetrating radar; image classification; image resolution; landmine detection; military radar; object detection; radar imaging; radar resolution; 3D ground penetrating radar image; GPR; VIR system; feature extraction; image classification algorithm; landmine detection; object detection; video impulse radar; Clutter; Ground penetrating radar; Image processing; Image resolution; Landmine detection; Object detection; Radar detection; Radar imaging; Radar measurements; Radar polarimetry;
Conference_Titel :
Microwaves, Radar and Wireless Communications, 2004. MIKON-2004. 15th International Conference on
Print_ISBN :
83-906662-7-8
DOI :
10.1109/MIKON.2004.1357056